Oscaf2 Contains Two CRM Domains and Is Necessary for Chloroplast

Shen et al. BMC Plant Biology (2020) 20:381 https://doi.org/10.1186/s12870-020-02593-z

RESEARCH ARTICLE Open Access OsCAF2 contains two CRM domains and is necessary for chloroplast development in rice Lan Shen1†, Qiang Zhang1†, Zhongwei Wang2, Hongling Wen1, Guanglian Hu1, Deyong Ren1, Jiang Hu1, Li Zhu1, Zhenyu Gao1, Guangheng Zhang1, Longbiao Guo1, Dali Zeng1 and Qian Qian1*

Abstract Background: Chloroplasts play an important role in plant growth and development. The chloroplast genome contains approximately twenty group II introns that are spliced due to proteins encoded by nuclear genes. CAF2 is one of these splicing factors that has been shown to splice group IIB introns in maize and Arabidopsis thaliana. However, the research of the OsCAF2 gene in rice is very little, and the effects of OsCAF2 genes on chloroplasts development are not well characterized. Results: In this study, oscaf2 mutants were obtained by editing the OsCAF2 gene in the Nipponbare variety of rice. Phenotypic analysis showed that mutations to OsCAF2 led to albino leaves at the seeding stage that eventually caused plant death, and oscaf2 mutant plants had fewer chloroplasts and damaged chloroplast structure. We speculated that OsCAF2 might participate in the splicing of group IIA and IIB introns, which differs from its orthologs in A. thaliana and maize. Through yeast two-hybrid experiments, we found that the C-terminal region of OsCAF2 interacted with OsCRS2 and formed an OsCAF2-OsCRS2 complex. In addition, the N-terminal region of OsCAF2 interacted with itself to form homodimers. Conclusion: Taken together, this study improved our understanding of the OsCAF2 protein, and revealed additional information about the molecular mechanism of OsCAF2 in regulating of chloroplast development in rice. Keywords: OsCAF2, CRM domain, Intron splicing, Chloroplast development, Rice

Background encoded polymerases (PEPs) and nucleus-encoded poly- Chloroplasts are necessary organelles for plant growth merases (NEPs) play key roles in regulating chloroplast and development, as they biosynthesize carbohydrates development. These studies suggest that expressions of through the fixation of CO2 [1, 2]. Chloroplasts are de- chloroplast genes require proteins encoded by the nu- rived from proplastids and are semiautonomous organ- clear genome [5, 6]. These proteins are involved in the elles [3]. The chloroplast genome is small and encodes transcription and post-transcription modification of approximately 100 proteins responsible largely for chloroplast related genes, such as chloroplast DNA repli- photosynthesis, carbon metabolism, and fatty acid syn- cation, RNA transcription, RNA editing, RNA stability, thesis [4]. Many studies have indicated that plastid- and RNA splicing [6]. The process of RNA splicing occurs by removing in- * Correspondence: [email protected] † trons from the precursor messenger mRNA (pre-mRNA) Lan Shen and Qiang Zhang contributed equally to this work. to join adjacent exons [7]. In plants, the introns of 1State Key Laboratory of Rice Biology / China National Rice Research Institute, Chinese Academy of Agricultural Sciences, Hangzhou 310006, China chloroplast genes are divided into two groups, known as Full list of author information is available at the end of the article

© The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Shen et al. BMC Plant Biology (2020) 20:381 Page 2 of 11

group I and group II introns, based on their conserved pri- complex, which was required for splicing of five subgroup mary and secondary structures, as well as the splicing IIB introns from the genes ndhA, ndhB, rps12, petB,and mechanism of RNA splicing. Group II introns are further ycf3 [23, 24]. The CFM2 has four CRM domains. In maize, divided into two subgroups, subgroup IIA and subgroup ZmCFM2 was required to splice the trnL, ndhA,andycf3 IIB [7, 8]. Previous work has shown that group I and introns [26]. In A. thaliana, AtCFM2 can splice the trnL, group II introns of chloroplast genes have lost the ability ndhA,andycf3 introns, and it influences splicing of the for autocatalytic splicing, and the introns of these genes clpP intron that was absent from the maize chloroplast lose elements that were necessary for RNA-based catalytic genome [26]. Unlike other proteins contained in the CRM centers [8, 9]. Chloroplast RNA splicing requires the re- domains, CFM3 contains three CRM domains and was lo- cruitment of proteins that are encoded by nuclear genes. cated in both chloroplasts and mitochondria. In A. thali- These proteins act as splicing factors that are essential for ana, AtCFM3a was necessary for splicing of the ndhB chloroplast RNA splicing reactions [7, 10]. In Arabidopsis intron [26]. In rice, fourteen proteins containing one or thaliana, chloroplast genome contains 20 group II introns more CRM domains have been identified [7]. The oscfm3 and 1 group I intron. In maize and rice, the chloroplast mutant resulted in an albino phenotype in seedlings genome consists of 17 group II introns and one group I caused by abnormal chloroplast development. OsCFM3 intron [11]. It has been reported that many nuclear genes promotes splicing of chloroplast genes ndhB, petD,and encode proteins that act as splicing factors that are in- rps16 [26]. Previous research showed that AL2 encodes volved in the regulation of chloroplast RNA splicing. For OsCRS1, which is homologous to AtCRS1 and ZmCRS1, example, pentatricopeptide repeat (PPR) proteins and and contains three CRM domains. However, OsCRS1 pro- chloroplast RNA splicing and ribosome maturation motes splicing of group I (trnL)andgroupIIintrons (CRM) domain proteins are involved in the regulation of (atpF, rpl2, ndhA, ndhB, petD,andycf3)thatdifferfrom splicing of chloroplast gene introns [7]. the homologous proteins AtCRS1 and ZmCRS1 [11]. It Previous studies also showed that many PPR proteins has been reported that OsCAF1 contains two CRM do- participate in the splicing of group II introns in various mains that could interact with OsCRS2 via its C-terminal plant species [12, 13]. For example, in maize, ZmPPR4 region and forms the OsCAF1-OsCRS2 complex in rice. and ZmPPR5 promote intron splicing of rps12 and trnG The OsCAF1 localizes to the chloroplast and affects spli- in the chloroplast [14, 15]. In A. thaliana, AtOTP70 and cing of subgroup IIA and subgroup IIB introns [27]. The AtOTP51 are required for intron splicing of rpoC1 and function of other proteins containing the CRM domain ycf3–2 [16, 17]. In rice, WSL4 and PGL12 participate in has not been studied extensively in rice. intron splicing of rpl2 and ndhA, respectively [18, 19]. In this study, we generated an oscaf2 mutant by using Inactivation of the PPR proteins involved in chloroplast the CRISPR/Cas9 (Clustered Regularly Interspaced Short intron splicing leads to abnormal chloroplast develop- Palindromic Repeats) gene-editing system and studied ment, with leaves showing chlorosis or albino seedling the role of OsCAF2 in rice chloroplast development. phenotypes [20, 21]. Our results showed that mutants of OsCAF2 lead to ab- With the exception of PPR proteins, some CRM normal chloroplast development and albino phenotypes domain-containing proteins act as splicing factors that in seedlings. Analysis of intron splicing of chloroplast regulate the splicing of chloroplast introns. It has been re- genes showed that OsCAF2 promoted the splicing of ported that the CRM domain-containing proteins influ- chloroplast subgroup IIA and subgroup IIB introns, ence chloroplast development in A. thaliana, maize, and which were necessary for chloroplast development. Simi- rice [7]. The first CRM gene, CRS1, was identified and lar to the ZmCAF2-ZmCRS2 complex, the C-terminal cloned in maize. ZmCRS1 contains three CRM domains, region of OsCAF2 interacted with OsCRS2 and formed which have high affinity and specificity in vitro for the the OsCAF2-OsCRS2 complex. Interestingly, OsCAF2 atpF intron, and they participate in intron splicing of atpF also could interact with itself through its N-terminal re- [9, 22]. ZmCAF1 and ZmCAF2 contain two CRM do- gion to form dimers that regulate chloroplast develop- mains, and each of them can interact with ZmCRS2 and ment in rice. These results demonstrate the molecular form ZmCAF1-ZmCRS2 and ZmCAF2-ZmCRS2 com- mechanism of OsCAF2 in rice chloroplast development. plexes [23]. The ZmCAF1-ZmCRS2 complex is required for the splicing of subgroup IIB introns, including rpl16, Results rps16, trnG, petD,andycf3 [24]. In A. thaliana, the Disruption of OsCAF2 function caused an albino AtCAF1-AtCRS2 complex also promoted the splicing of phenotype the rpoC1 and ClpP introns, which were absent in the According to previous studies, the gene homologous to maize chloroplast genome [25]. Furthermore, earlier stud- ZmCAF2 and AtCAF2 from maize and A. thaliana was iden- ies showed that the C-terminal region of ZmCAF2 can tifiedfromthericegenomeandnamedOsCAF2 (LOC_ interact with ZmCRS2 to form the ZmCAF2-ZmCRS2 Os01g21990)[7]. The OsCAF2 contains six exons, with a Shen et al. BMC Plant Biology (2020) 20:381 Page 3 of 11

total coding region of 1824 bp, which encodes 608 amino Expression analysis of chloroplast development and acids. The OsCAF2 contains two CRM domains, but their photosynthesis-related genes in oscaf2 mutants functions are unknown. To study the function of OsCAF2, To further analysis the effects of impaired chloroplast we constructed the OsCAF2 gene editing vector including development in the oscaf2 mutant, we measured the ex- two target sgRNAs in exon 1 (upstream of the two CRM do- pression levels of chloroplast development and mains) and exon 3 (the first CRM domain) respectively, and photosynthesis-related genes. PEPs and NEPs are regu- transformed it into Nipponbare via the Agrobacterium lated transcriptionally by chloroplast-encoded genes re- method (Fig. 1a-b). In the T0 generation, we obtained 15 sponsible for chloroplast development. Expression levels plants. Through PCR amplification and sequencing of of the NEP component gene OsRpoTp in the oscaf2 mu- OsCAF2, we screened one homozygous mutants and one tant was normal compared to the WT (Fig. 2c). How- heterozygote plant, named oscaf2 and oscaf2/OsCAF2, re- ever, except the gene rpoA, there were significant spectively (Fig. 1c). The mutation types of one or two base increases in the expression levels of PEP component insertions lead to the premature termination of OsCAF2 genes, including rpoB, rpoC1, and rpoC2, in the oscaf2 translation. Phenotypic observations suggested that the mutant relative to WT (Fig. 2c). By contrast, the expres- oscaf2 mutants showed albino seedling phenotypes and died sions of photosynthesis-related genes involving RbcS, at the three leaf stage, but the heterozygote oscaf2/OsCAF2 RbcL, HEMA, CHLI, psbO, HEMD, psbA, and CHLM showed green leaves and normal growth (Fig. 1d). To investi- were dramatically reduced in the oscaf2 mutant (Fig. 2c). gate the function of OsCAF2, we obtained oscaf2 mutants in These results suggest that OsCAF2 might affect the ex- the T1 generation of oscaf2/OsCAF2 heterozygote plant pression level of chloroplast development and (Fig. 2a). The oscaf2 mutants from oscaf2/OsCAF2 heterozy- photosynthesis-related genes. gote plant also appeared as albino and then died at the three leaf stage (Fig. 2a). Compared to wild type (WT), the Chl a, Chloroplast development was defected in oscaf2 mutant Chl b, car and Chl content in oscaf2 mutant leaves To further study the albino leaf phenotype in the oscaf2 were significantly decreased, and was practically de- mutant, we observed the ultrastructure of chloroplasts of void of car and Chl b (Fig. 2b). These results suggest leaves in WT and oscaf2 in seedling stage using trans- that OsCAF2 was necessary for chloroplast develop- mission electron microscopy (TEM). Our result showed ment, and that mutation of OsCAF2 resulted in an that the integrated chloroplast structures and normal albino phenotype in seedlings. grana stacks were found in WT leaves (Fig. 2d-e), and

Fig. 1 Disruption of OsCAF2 results in an albino phenotype. a Schematic diagram of CRISPR/Cas9 system for editing OsCAF2. b Diagram of the

targeted site in OsCAF2. c Mutation types of transformed plants in the T0 generation. The protospacer adjacent motif sequence is shown in red. The targeted sequences are shown in blue. d The phenotype of the two positive plants at the seedling stage. Scale bar = 1 cm Shen et al. BMC Plant Biology (2020) 20:381 Page 4 of 11

Fig. 2 Chloroplast development was impaired in the oscaf2 mutant. a Phenotype presentation of WT (#1) and oscaf2 mutant (#2 and #3). Scale bar = 1 cm. b Chlorophyll contents of WT and oscaf2 at the two leaf stage. c Expression analysis of chloroplast synthesis and photosynthesis- related genes in WT and the oscaf2 mutant. The values represent the mean of three independent experiments. Error bars indicate SE, ** indicate P < 0.01, respectively. d-g The chloroplast ultrastructure of WT and oscaf2 mutants from leaves. WT (d, e) and oscaf2 mutants (f, g). CP, chloroplast, GL, grana lamella. Scale bars of d and f,2μm, Scale bars of e and g,1μm Shen et al. BMC Plant Biology (2020) 20:381 Page 5 of 11

the abnormal chloroplast structure and the disorganized Expression patterns and subcellular localization of grana lamellar stacks in oscaf2 mutants (Fig. 2f-g). These OsCAF2 data suggest that the mutation to OsCAF2 impaired chlo- According to the gene expression and protein tools of roplasts development in rice. These results also suggest rice (http://www.bar.utoronto.ca/efprice/cgi-bin/efpWeb. that OsCAF2 is necessary for chloroplast development. cgi), OsCAF2 was highly expressed in leaves but low or undetectable level was found in spikes and root. To val- idate the result, we analyzed the expression pattern of OsCAF2 influences splicing of group II introns OsCAF2 in different tissues at the heading stage of Nip- In A. thaliana and maize, CAF2 promoted the splicing ponbare using qRT-PCR. Compared with the root, the of the chloroplast RNA group IIB introns [25]. There- expression level of other tissues were significantly in- fore, we examined whether OsCAF2 was involved in creased, and the gene expression in leaves even more chloroplast RNA intron splicing in rice. The rice chloro- than a thousand times (Fig. 4a). Our results indicated plast genome contains 17 group II introns and 1 group I that OsCAF2 expression was highest in green tissues, es- intron. We amplified chloroplast genes that contained at pecially in leaves. least one intron using RT-PCR and compared intron The Wolf PSORT (https://wolfpsort.hgc.jp/) website splicing efficiency between the WT and oscaf2 mutant. predicted that OsCAF2 was localized to rice chloro- We found abnormal intron splicing of subgroup IIA in- plasts. To determine the subcellular localization of trons (rpl2, rps12, and atpF) and subgroup IIB introns OsCAF2, we constructed the OsCAF2-GFP fusion vector (ndhA, ndhB, and ycf3) in the oscaf2 mutant, compared and transformed it into rice protoplasts. The p580 empty with the WTs (Fig. 3, Fig. S1). However, the intron spli- vector was used as a control. The results showed that cing of trnL (group I) was not affected in the oscaf2 mu- the empty vector had green fluorescent signals in the tant (Fig. 3, Fig. S1). These results suggested that nucleus, membrane, and cytoplasm, but the OsCAF2- OsCAF2 might participate in splicing of subgroup IIA GFP fusion protein produced signals that co-localized and subgroup IIB introns but not group I introns. with the chloroplast autofluorescence signals, suggesting

Fig. 3 Splicing analysis of group I and group II introns in WT and oscaf2 mutants. U indicates unspliced transcripts. S indicates spliced transcripts Shen et al. BMC Plant Biology (2020) 20:381 Page 6 of 11

Fig. 4 Expression pattern and subcellular localization of OsCAF2. a Expression pattern of OsCAF2 in various tissues, including root, stem, leaf, sheath, and spike at the heading stage. The values represent the mean of three independent experiments. Error bars indicate SE, ** indicates P < 0.01. b Subcellular localization of OsCAF2 protein in rice protoplasts that OsCAF2 localizes in chloroplasts (Fig. 4b, Fig. S2). constructed OsCAF2-AD and OsCAF2-BD vectors and This result supports the view that OsCAF2 plays an im- then co-transformed them into AH109 yeast cells. The portant role in regulating chloroplast development. Y2H results showed that OsCAF2 could self-interact and form homodimers (Fig. 6a, Fig. S4). We divided OsCAF2 C-terminal region of OsCAF2 interacts with OsCRS2 into the three sections as above, to further analyze which Previous studies show that the N-terminal region of CAF2 section was responsible for forming homodimers. Only the interacts with CRS2 and forms the CAF2-CRS2 complex in N-terminal region of OsCAF2 was able to interact with it- maize [23]. To confirm the relationship between CAF2 and self, but it could not interact either with the M or C- CRS2 in rice, we acquired the coding sequence (CDS) of terminal regions of OsCAF2 (Fig. 6b, Fig. S4). Therefore, OsCRS2 (LOC_Os01g04130) from rice leaves and con- the N-terminal of OsCAF2 was necessary to form structed the OsCRS2-PGADT7 (OsCRS2-AD) and homodimers. OsCAF2-PGBKT7 (OsCAF2-BD) vectors for yeast two- hybrid (Y2H) analysis. The AH109 yeast cell that contained Discussion either OsCRS2-AD and the empty BD vector or OsCAF2- The OsCAF2 protein has two CRM domains with hom- BD and the empty AD vector were able to grow on syn- ology to AtCAF2 and ZmCAF2. Our phenotypic results thetic dextrose medium lacking Leu and Trp (SD-T/L), but of the oscaf2 mutant were consistent with the albino they were unable to grow on medium lacking Leu, Trp, mutant atcaf2 in A. thaliana, which was obtained by T- His, and Ade (SD-L/T/H/A). However, the AH109 yeast DNA insertion [25]. Therefore, we suspected that the cell that contained OsCRS2-AD and OsCAF2-BD, function of CAF2 might be highly conserved in A. thali- OsCRS2-BD and OsCAF2-AD were able to grow in SD-T/ ana and rice. In rice, AL2, OsCAF1, and CFM3 encoding L and SD-L/T/H/A (Fig. 5a, Fig. S3). These results suggest proteins contained CRM domains, and the al2, oscaf1, that OsCAF2 can interact with OsCRS2 to form the and oscfm3 mutants showed albino leaf phenotypes with OsCAF2-OsCRS2 complex in rice. We then truncated the disrupted chloroplast structure [11, 26, 27]. Research OsCAF2 protein into three sections, named OsCAF2-N shows there were 14 proteins containing one or more (N-terminal), OsCAF2-M (middle), and OsCAF2-C (C-ter- CRM domains, and four of them contained two CRM minal) and repeated the Y2H experiment (Fig. 5b). Among domains in rice [7]. As the albino phenotypic of oscaf2 these sections, OsCAF2-M contains the two CRM domains. mutants, we speculated that the function of OsCAF2 The OsCAF2-C interacted with OsCRS2 but OsCAF2-N was not redundant with the other three proteins that and OsCAF2-M did not (Fig. 5c, Fig. S3). These results in- contain two CRM domains. These results suggest that dicate that the C-terminal of OsCAF2 is necessary for the CRM domain proteins play a key role in chloroplast de- interaction with OsCRS2. velopment in rice. It has been reported that CAF2 is a splicing factor that OsCAF2 forms homodimers through the N-terminal is encoded by nuclear genes that participate in intron region splicing of petB, rps12, ndhA, and ycf3 in maize and A. In order to study whether OsCAF2 could form homodi- thaliana [25]. This indicates that the molecular mechan- mers, we carried out an Y2H analyzes. In brief, we ism of CAF2 regulation of chloroplast development is Shen et al. BMC Plant Biology (2020) 20:381 Page 7 of 11

Fig. 5 OsCAF2 interacts with OsCRS2 through the C-terminal region. a The Y2H interaction between OsCAF2 and OsCRS2. The full-length CDS of OsCAF2 and OsCRS2 were cloned into pGBKT7 (BD) and pGADT7 (AD) vectors to produce OsCAF2-BD, OsCAF2-AD, OsCRS2-BD, and OsCRS2-AD plasmids, respectively. The OsCAF2-BD and OsCRS2-AD, OsCAF2-AD and OsCRS2-BD were co-transformed into the AH109 yeast strain. The OsCAF2-BD and AD plasmid, and OsCRS2-AD and BD plasmid were co-transformed into the AH109 yeast strain as a control. b Schematic diagram of OsCAF2 protein structure. Different sections of OsCAF2 involving in OsCAF2-N (1–230 aa), OsCAF2-M (231–430 aa), and OsCAF2-C (431–608 aa). c Y2H interaction of OsCRS2 and different truncated sections of OsCAF2. The truncated sections of OsCAF2 were cloned into the BD vector to generate OsCAF2-N-BD, OsCAF2-M-BD, and OsCAF2-C-BD vectors. These three vectors were co-transferred into AH109 yeast cells with the OsCRS2-AD vector. The yeast cells were cultured on SD-T/L and SD−/T/L/H/A media with different dilution series (1, 10− 1, and 10− 2) relatively conserved in maize and A. thaliana. In this AtCAF2 and ZmCAF2. Similar to OsCAF1 and AL2 in study, by analyzing the intron splicing of chloroplast rice, OsCAF2 affects splicing of subgroup IIA and sub- genes, we found that OsCAF2 promotes the splicing of group IIB introns [11, 27]. The molecular mechanism of subgroup IIB and subgroup IIA introns. For example, OsCAF2 regulating chloroplast development may be dif- OsCAF2 influences rpl2 and atpF intron splicing, but ferent from AtCAF2 and ZmCAF2. The function of ZmCAF2 and AtCAF2 does not influence splicing of OsCAF2 in splicing of subgroup IIA introns, such as these two introns in maize and A. thaliana. However, atpF and rpl2 introns, requires further study. OsCAF2, AtCAF2, and ZmCAF2 all affect splicing of In plants, chloroplast development is mainly divided ndhA and ycf3 introns. Therefore, this suggests that into three steps [28]. NEP and PEP participate in the OsCAF2 has a conserved but distinct role compared to second and third steps to regulate the transcription of Shen et al. BMC Plant Biology (2020) 20:381 Page 8 of 11

Fig. 6 OsCAF2 interacts with itself through the N-terminal region. a Y2H interaction of OsCAF2 with itself. The full-length CDS of OsCAF2 was cloned into BD and AD vectors, to produce OsCAF2-BD and OsCAF2-AD vectors, respectively. The OsCAF2-BD and OsCAF2-AD were co- transformed into the AH109 yeast strain. The OsCAF2-BD and AD plasmid, and the OsCAF2-AD and BD plasmid were co-transformed into the AH109 yeast strain as control. b Y2H interaction of OsCAF2 with itself. The truncated sections of OsCAF2 were cloned into BD and AD vectors to produce the OsCAF2-N-BD, OsCAF2-M-BD, OsCAF2-C-BD, OsCAF2-N-AD, OsCAF2-M-AD and OsCAF2-C-AD vectors. A pair of AD and BD was co- transformed into the AH109 yeast strain. The yeast cells were cultured on SD-T/L and SD−/T/L/H/A media with different dilution series (1, 10− 1, and 10− 2) plastid genes [28]. In the oscaf2 mutant, the expression RbcL, HEMA, CHLI, psbO, HEMD, psbA, and CHLM de- levels of the NEP component (OsRpoTp) are similar to creased. It may be possible that the signals from the the WT, while the expression of PEP components (rpoB, plasmid to the nucleus might be different in the oscaf2 rpoC1, and rpoC2) increased. This might be caused by a mutant. This result of signal transduction was also con- feedback regulation mechanism: impaired chloroplast sistent with previous results in the osptac2 mutant [30]. development might induce transcription of genes with Thus, these results showed that OsCAF2 might influence PEP components. Similar conclusions were acquired in chloroplast development through its effects on PEP and many chloroplast defective mutants, such as asl3 and photosynthesis-related gene expression. sal4, in rice. Previous work showed that asl3 and sal4 It was reported that CAF2 interacted with CRS2 also exhibited albino leaf phenotypes and expression through its C-terminal region in maize [23]. CRS2 was a levels of rpoB, rpoC1, and rpoC2 in asl3 and sal4 in- group II intron splicing factor that could promote the creased significantly [18, 29]. By contrast, expression splicing of nine group II introns [31, 32]. Gene expres- levels of photosynthesis-related genes, such as RbcS, sion analysis showed that expression levels of OsCAF2 Shen et al. BMC Plant Biology (2020) 20:381 Page 9 of 11

were highest in leaves. The expression pattern of provided by the China National Rice Research Institute OsCAF2 was similar to OsCRS2. In this study, we deter- (CNRRI) in Hangzhou, Zhejiang Province, China. We mined that OsCAF2 interacts with OsCRS2 through its designed the two target sites of OsCAF2 (GTGCGGCT C-terminal region. Previous analysis of the C-terminal GCACACCCCGCTCGG, AAGGTTATTCACCGAG region of CAF2 in monocot and dicot plants, including TAGGTGG) and constructed the pC1300-Ubi::Cas9- Arabidopsis thaliana, maize and rice, found that there gXOsCAF2-g1-gXOsCAF2-g2 vector, according to previously- was a conservative residue of CAF2 that formed a hydro- described method [34]. Vector construction and sequen- phobic region necessary for protein interactions with cing related primers are listed in Table S1. This vector CRS2 [23]. In rice, OsCAF2 might interact with OsCRS2 were transformed into Nipponbare via Agrobacterium- through this conservative residue. The molecular mech- mediated transformation. The OsCAF2 mutation of the anism of the CAF2-CRS2 complex formation might be positive transgenic plants were identified by the Sanger conserved across plants. It has been reported that the method and analyzed by the degenerate sequence decod- homodimers and heterodimers might have biological ad- ing method [35]. All the transgenic plants we obtained vantages than the individual protein, which could pro- and their offspring were grown in the greenhouse under mote the synthesis of products or regulate the activity of constant temperature (30 °C at 16 h light and 8 h dark) partner enzyme domains [33]. Consider the homodimers of CNRRI in Hangzhou, Zhejiang Province, China. of OsCAF2 formed via its N-terminal region, we suspected that proper chloroplast development requires recruitment of OsCAF2 and OsCRS2 to form the Pigment content measurement OsCAF2-OsCRS2 complex, as well homodimers of The Chlorophyll a (Chla), chlorophyll b (Chlb), caroten- OsCAF2 to achieve splicing of group II introns. Previous oid (Car) and total chlorophyll (Chl) content was mea- studies have shown that OsCAF1 also contains two sured. In brief, we obtained WT and oscaf2 mutant CRM domains that could affect splicing of subgroup IIA leaves (0.2 g) at the seedling stage. The leaves were cut and subgroup IIB introns, and the orthologous genes of and soaked in 5 ml 95% ethanol for 48 h under dark con- AtCAF1 and ZmCAF1 in A. thaliana and maize only ditions. Spectrophotometric measurements were con- promoted the splicing of subgroup IIB introns [27]. In ducted using a UV-1800PC spectrophotometer (Mapada, addition, OsCAF1 interacts with itself to form homodi- China) at 470 nm, 649 nm, and 665 nm. According to mers. In maize, ZmCRS1 contains three CRM domains previous methods [36], we calculated the Chla, Chlb, and can form homodimers, but whether CAF2 forms Car and Chl content in WT and oscaf2 mutant leaves. homodimers has not been reported in maize and A. thaliana [22]. Therefore, this study provides new clues Transmission electron microscopy (TEM) assays to the function of the CAF2 protein in different plants At the three leaf stage, we samples of WT and oscaf2 and provides important information for exploring the mutant leaves were cut into 0.1 cm × 0.1 cm pieces. The development of chloroplasts in rice. samples were fixed in 2.5% glutaraldehyde for 4 h in at 4 °C (pH 7.0) and then fixed in 1% OsO4 (pH 7.4). The Conclusions fixed samples were stained with uranyl acetate. After de- The present study generated albino lethal oscaf2 mutants hydration in an ethanol series, the samples were embed- using CRISPR/Cas9 gene-editing system. The following ded in acrylic resin. The samples were staining and study showed that OsCAF2 could affect chloroplast devel- observed with a Hitachi-7500 (Tokyo, Japan) transmis- opment, and might participate in the splicing of group IIA sion electron microscope. and IIB introns, which differs from its orthologs in A. thaliana and maize. The Y2H assay showed that the C- terminal region of OsCAF2 interacted with OsCRS2 and RNA extraction and quantitative real-time PCR (qRT-PCR) formed an OsCAF2-OsCRS2 complex, and the N- Total RNA was extracted from leaves using RNA extrac- terminal region of OsCAF2 interacted with itself to form tion kit (Axygen, USA) according to the manufacturer’s homodimers. These findings improved our understanding protocols. First-strand cDNA was synthesized using the of the OsCAF2 protein, and revealed additional informa- ReverTra Ace qPCR RT Kit (TOYOBO, Japan) accord- tion about the molecular mechanism of OsCAF2 in regu- ing to the manufacturer’s instructions. The qRT-PCR lating of chloroplast development in rice. analysis process was as follows: 5 min at 94 °C followed by 40 cycles of 95 °C for 15 s and 58 °C for 50 s. The Methods OsActin1 was used as a reference gene in this study. The Plant materials and growth conditions relative expression levels were calculated according to −ΔΔ The japonica variety Nipponbare have been convention- the 2 CT method [37]. All primers for qRT-PCR are ally cultivated in Japan. The seeds of Nipponbare were shown in Table S1. Shen et al. BMC Plant Biology (2020) 20:381 Page 10 of 11

Splicing analysis of group I and group II introns Funding We used RT-PCR to perform splicing analysis of group I This research was supported by the National Natural Science Foundation of China (No. 31801333, No. 31701390), and the Central Public-interest Scientific and group II introns of chloroplast genes from WT and Institution Basal Research Fund of China National Rice Research Institute oscaf2 mutants. The RT-PCR procedure was as follows: (2017RG001–4). The funding bodies provided the financial support to this re- 94 °C for 5 min, followed by 30 cycles of 94 °C for 30 s, search, including experimental design and implementation, sampling and data analysis. No funder played the role in data collection and analysis and 58 °C for 30 s, 72 °C for 45 s, and a final elongation step writing the manuscript. at 72 °C for 10 min. All chloroplast RNA splicing analysis primers are listed in Table S1. Availability of data and materials The datasets supporting the results of this article are available from the corresponding author on reasonable request. Sequence data used during Subcellular localization of OsCAF2 the current study for the cDNA and genomic DNA of OsCAF2 and OsCRS2 To confirm subcellular localization of the OsCAF2 protein, are available from the GenBank/EMBL data libraries under accession numbers LOC_Os01g21990, LOC_Os01g04130, respectively, and could also available the full-length OsCAF2 coding sequence (remove stop from the National Center for Biotechnology Information (NCBI) (OsCAF2 code) was amplified and cloned into the pAN580 vector to Gene ID: 4326994, OsCRS2 Gene ID: 4325801). create the p580–2 × CaMV35S::OsCAF2-GFP (OsCAF2- GFP) plasmid (Table S2). The OsCAF2-GFP vector con- Ethics approval and consent to participate struction primers are listed in Table S1. The preparation of Not applicable. rice protoplasts and transformation followed previously de- Consent for publication scribed methods [38]. The fluorescence of green fluorescent Not applicable. protein (GFP) was observed using confocal laser scanning microscopy (Zeiss LSM 780, Germany). Competing interests The authors declare that they have no competing interests.

Yeast two-hybrid (Y2H) assays Author details 1 The full-length and truncated coding sequences of State Key Laboratory of Rice Biology / China National Rice Research Institute, Chinese Academy of Agricultural Sciences, Hangzhou 310006, OsCAF2, and full-length OsCRS2 coding sequence were China. 2Biotechnology Research Center, Chongqing Academy of Agricultural amplified via PCR (Table S2). Relevant PCR amplification Sciences, Chongqing 401329, China. primers are listed in Table S1. PCR amplification products Received: 18 March 2020 Accepted: 12 August 2020 were cloned into the pGAD-T7 or pGBK-T7 vectors. The Y2H experiment used the Clontech two-hybrid system, according to the manufacturer’s instructions. References 1. Jarvis P, López-Juez E. Biogenesis and homeostasis of chloroplasts and other plastids. Nat Rev Mol Cell Biol. 2013;14(12):787–802. Supplementary information 2. Li Y, Zhang J, Li L, Gao L, Xu J, Yang M. Structural and comparative analysis Supplementary information accompanies this paper at https://doi.org/10. of the complete chloroplast genome of Pyrus hopeiensis- “wild plants with a 1186/s12870-020-02593-z. tiny population” -and three other Pyrus species. Int J Mol Sci. 2018;19(10): 3262. 3. Bobik K, Burch-Smith TM. Chloroplast signaling within, between and beyond Additional file 1: Table S1. The primers used in present study. Table cells. Front Plant Sci. 2015;6:781. S2. Sequence of OsCAF2 and OsCRS2 genes from Nipponbare. 4. Nakai M. New perspectives on chloroplast protein import. Plant Cell Physiol. Additional file 2: Figure S1. Original images for Fig. 3. 2018;59(6):1111–9. Additional file 3: Figure S2. Original images for Fig. 4b. 5. Hedtke B, Börner T, Weihe A. Mitochondrial and chloroplast phage-type RNA polymerases in Arabidopsis. Science. 1997;277(5327):809–11. Additional file 4: Figure S3. Original images for Fig. 5. 6. Yu QB, Huang C, Yang ZN. Nuclear-encoded factors associated with the Additional file 5: Figure S4. Original images for Fig. 6. chloroplast transcription machinery of higher plants. Front Plant Sci. 2014;5: 316. 7. De Longevialle AF, Small ID, Lurin C. Nuclearly encoded splicing factors Abbreviations implicated in RNA splicing in higher plant organelles. Mol Plant. 2010;3(4): PEPs: Plastid-encoded polymerases; NEPs: Nucleus-encoded polymerases; 691–705. pre-mRNA: Precursor messenger mRNA; PPR: Pentatricopeptide repeat; 8. Bonen L, Vogel J. The ins and outs of group II introns. Trends Genet. 2001; CRM: Chloroplast RNA splicing and ribosome maturation; CRISPR: Clustered 17(6):322–31. regularly interspaced short palindromic repeats; TEM: Transmission electron 9. Till B, Schmitz-Linneweber C, Williams-Carrier R, Barkan A. CRS1 is a novel microscopy; Y2H: Yeast two-hybrid; qRT-PCR: RNA extraction and group II intron splicing factor that was derived from a domain of ancient Quantitative real-time PCR; GFP: Green fluorescent protein; CDS: Coding origin. Rna. 2001;7(9):1227–38. sequence 10. Barkan A, Goldschmidt-Clermont M. Participation of nuclear genes in chloroplast gene expression. Biochimie. 2000;82(6–7):559–72. Acknowledgements 11. Liu C, Zhu H, Xing Y, Tan J, Chen X, Zhang J, et al. Albino Leaf 2 is involved Not applicable. in the splicing of chloroplast group I and II introns in rice. J Exp Bot. 2016; 67(18):5339–47. Authors’ contributions 12. Khrouchtchova A, Monde RA, Barkan A. A short PPR protein required for the SL and ZQ designed the study and wrote the original manuscript. ZQ, SL, splicing of specific group II introns in angiosperm chloroplasts. Rna. 2012; WZW, WHL, and HGL conducted all the experiments. RDY, HJ, ZL, GZY, ZGH 18(6):1197–209. and GLB were responsible for the data analysis ZDL and QQ modified the 13. Rovira AG, Smith AG. PPR proteins - orchestrators of organelle RNA manuscript. All authors read and approved the manuscript. metabolism. Physiol Plant. 2019;166(1):451–9. Shen et al. BMC Plant Biology (2020) 20:381 Page 11 of 11

14. Schmitz-Linneweber C, Williams-Carrier RE, Williams-Voelker PM, Kroeger TS, 37. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using Vichas A, Barkan A. A pentatricopeptide repeat protein facilitates the trans- real-time quantitative PCR and the 2−ΔΔCT method. Methods. 2001;25(4): splicing of the maize chloroplast rps12 pre-mRNA. Plant Cell. 2006;18(10): 402–8. 2650–63. 38. Zhang Y, Su J, Duan S, Ao Y, Dai J, Liu J, et al. A highly efficient rice green 15. Tadini L, Ferrari R, Lehniger MK, Mizzotti C, Moratti F, Resentini F, et al. tissue protoplast system for transient gene expression and studying light/ Trans-splicing of plastid rps12 transcripts, mediated by AtPPR4, is essential chloroplast-related processes. Plant Methods. 2011;7(1):30. for embryo patterning in Arabidopsis thaliana. Planta. 2018;248(1):257–65. 16. De Longevialle AF, Hendrickson L, Taylor NL, Delannoy E, Lurin C, Badger M, ’ et al. The pentatricopeptide repeat gene OTP51 with two LAGLIDADG Publisher sNote motifs is required for the cis-splicing of plastid ycf3 intron 2 in Arabidopsis Springer Nature remains neutral with regard to jurisdictional claims in thaliana. Plant J. 2008;56(1):157–68. published maps and institutional affiliations. 17. Chateigner-Boutin AL, Des Francs-Small CC, Delannoy E, Kahlau S, Tanz SK, De Longevialle AF, et al. OTP70 is a pentatricopeptide repeat protein of the E subgroup involved in splicing of the plastid transcript rpoC1. Plant J. 2011; 65(4):532–42. 18. Wang Z, Lv J, Xie S, Zhang Y, Qiu Z, Chen P, et al. OsSLA4 encodes a pentatricopeptide repeat protein essential for early chloroplast development and seedling growth in rice. Plant Growth Regul. 2018;84:249–60. 19. Chen L, Huang L, Dai L, Gao Y, Zou W, Lu X, et al. PALE-GREEN LEAF12 encodes a novel Pentatricopeptide repeat protein required for chloroplast development and 16S rRNA processing in Rice. Plant Cell Physiol. 2019; 60(3):587–98. 20. Beick S, Schmitz-Linneweber C, Williams-Carrier R, Jensen B, Barkan A. The pentatricopeptide repeat protein PPR5 stabilizes a specific tRNA precursor in maize chloroplasts. Mol Cell Biol. 2008;28(17):5337–47. 21. Wang Y, Ren Y, Zhou K, Liu L, Wang J, Xu Y, et al. WHITE STRIPE LEAF4 encodes a novel P-type PPR protein required for chloroplast biogenesis during early LEAF development. Front Plant Sci. 2017;8:1116. 22. Ostersetzer O, Cooke AM, Watkins KP, Barkan A. CRS1, a chloroplast group II intron splicing factor, promotes intron folding through specific interactions with two intron domains. Plant Cell. 2005;17(1):241–55. 23. Ostheimer GJ, Rojas M, Hadjivassiliou H, Barkan A. Formation of the CRS2- CAF2 group II intron splicing complex is mediated by a 22-amino acid motif in the COOH-terminal region of CAF2. J Biol Chem. 2006;281(8):4732– 8. 24. Ostheimer GJ, Williams-Carrier R, Belcher S, Osborne E, Gierke J, Barkan A. Group II intron splicing factors derived by diversification of an ancient RNA- binding domain. EMBO J. 2003;22(15):3919–29. 25. Asakura Y, Barkan A. Arabidopsis orthologs of maize chloroplast splicing factors promote splicing of orthologous and species-specific group II introns. Plant Physiol. 2006;142(4):1656–63. 26. Asakura Y, Bayraktar OA, Barkan A. Two CRM protein subfamilies cooperate in the splicing of group IIB introns in chloroplasts. Rna. 2008;14(11):2319–32. 27. Zhang Q, Shen L, Wang Z, Hu G, Ren D, Hu J, et al. OsCAF1, a CRM domain containing protein, influences chloroplast development. Int J Mol Sci. 2019; 20:18. 28. Zhang Z, Cui X, Wang Y, Wu J, Gu X, Lu T. The RNA editing factor WSP1 is essential for chloroplast development in Rice. Mol Plant. 2017;10(1):86–98. 29. Lin D, Gong X, Jiang Q, Zheng K, Zhou H, Xu J, et al. The rice ALS3 encoding a novel pentatricopeptide repeat protein is required for chloroplast development and seedling growth. Rice (New York, NY). 2015;8:17. 30. Wang D, Liu H, Zhai G, Wang L, Shao J, Tao Y. OspTAC2 encodes a pentatricopeptide repeat protein and regulates rice chloroplast development. J Genet Genomics. 2016;43(10):601–8. 31. Jenkins BD, Kulhanek DJ, Barkan A. Nuclear mutations that block group II RNA splicing in maize chloroplasts reveal several intron classes with distinct requirements for splicing factors. Plant Cell. 1997;9(3):283–96. 32. Jenkins BD, Barkan A. Recruitment of a peptidyl-tRNA hydrolase as a facilitator of group II intron splicing in chloroplasts. EMBO J. 2001;20(4): 872–9. 33. Panicot M, Minguet EG, Ferrando A, Alcázar R, Blázquez MA, Carbonell J, et al. A polyamine metabolon involving aminopropyl transferase complexes in Arabidopsis. Plant Cell. 2002;14(10):2539–51. 34. Wang C, Shen L, Fu Y, Yan C, Wang K. A simple CRISPR/Cas9 system for multiplex genome editing in Rice. J Genet Genom. 2015;42(12):703–6. 35. Ma X, Chen L, Zhu Q, Chen Y, Liu Y. Rapid decoding of sequence-specific nuclease-induced heterozygous and Biallelic mutations by direct sequencing of PCR products. Mol Plant. 2015;8(8):1285–7. 36. Lichtenthaler H. Chlorophylls and carotenoids: pigments of photosynthetic biomembranes. Methods Enzymol. 1987;148:350–82.